Alan L. Yuille

Artists can now directly sculpt 3D object motion relative to a moving camera, overcoming the inherent ambiguity of 2D controls in video generation.

Routine clinical reports and historical scans can entirely replace manual voxel annotations, yielding dense tumor segmenters that outperform supervised public baselines at zero labeling cost.

Models trained with VBVR-Pro not only excel in a controlled task space but also show significant transferability to external benchmarks, revealing critical insights into visual reasoning mechanisms.